Operations

Management Chapter 3
PROF. DR. FRANCESCO D. SANDULLI
UNIVERSITY COMPLUTENSE OF MADRID
Operations Management Chapter 3
Capacity
Location
Capacity
► The throughput, or the number of units
a facility can hold, receive, store, or
produce in a period of time
► Determines fixed costs
► Determines if demand will be satisfied
► Three time horizons

https://www.youtube.com/watch?v=8SqnC8ZPgnY
Capacity
Time Horizon
Options for Adjusting Capacity

Long-range Add facilities

planning
Intermediate-
Add long lead time equipment *
range Subcontract Add personnel
planning Add equipment Build or use inventory
(aggregate Add shifts
planning)
Schedule jobs
Short-range Schedule personnel
planning
(scheduling)
* Allocate machinery

Modify capacity Use capacity

* Difficult to adjust capacity as limited options exist
Capacity
►Design capacity is the maximum theoretical output of a
system
► Normally expressed as a rate

►Effective capacity is the capacity a firm expects to achieve

given current operating constraints
► Often lower than design capacity
Capacity
Utilization is the percent of design
capacity actually achieved
Utilization = Actual output/Design capacity

Efficiency is the percent of effective

capacity actually achieved
Efficiency = Actual output/Effective capacity
Capacity
Actual production last week = 148,000 rolls
Effective capacity = 175,000 rolls
Design capacity = 1,200 rolls per hour
Bakery operates 7 days/week, 3 - 8 hour shifts

Design capacity = (7 x 3 x 8) x (1,200) = 201,600 rolls

Utilization = 148,000/201,600 = 73.4%

Efficiency = 148,000/175,000 = 84.6%

Bottleneck Analysis and the
Theory of Constraints
► The bottleneck time is the time of the slowest workstation
(the one that takes the longest) in a production system
►The throughput time is the time it takes a unit to go through
production from start to end

A B C

2 min/unit 4 min/unit 3 min/unit

Capacity Analysis
►Two identical sandwich lines
►Lines have two workers and three operations
►All completed sandwiches are wrapped

Bread Fill
15 sec/sandwich 20 sec/sandwich
Wrap/
Order Toaster
Deliver
30 sec/sandwich 20 sec/sandwich
Bread Fill 37.5 sec/sandwich

15 sec/sandwich 20 sec/sandwich
Capacity Order
Bread
15 sec
Fill
20 sec
Toaster
Wrap/
Deliver

Analysis 30 sec
Bread
15 sec
Fill
20 sec
20 sec
37.5 sec

►The two lines each deliver a sandwich every 20 seconds

At 37.5 seconds, wrapping and delivery has the longest
►
processing time and is the bottleneck
Capacity per hour is 3,600 seconds/37.5 seconds/sandwich = 96
►
sandwiches per hour
►Throughput time is 30 + 15 + 20 + 20 + 37.5 = 122.5 seconds
Theory of Constraints
►Five-step process for recognizing and managing limitations

Step 1: Identify the constraints

Step 2: Develop a plan for overcoming the constraints
Step 3: Focus resources on accomplishing Step 2
Step 4: Reduce the effects of constraints by offloading
work or expanding capability
Step 5: Once overcome, go back to Step 1 and find
new constraints
Bottleneck Management
1. Release work orders to the system at the pace set by the
bottleneck
► Drum: Product time of the Bottleneck
► Buffer:Period of time to protect the drum resource from problems that
occur upstream from the drum operation. Its effect to provide a
resynchronization of the work as it flows through the plant.
► Rope: After the drum has been scheduled, material release and
shipping are connected to it, using the buffer offset. Material is
released at the same rate as the drum can consume it. Orders are
shipped at the rate of drum production.

2. Lost time at the bottleneck represents lost time for the whole
system
3. Increasing the capacity of a non-bottleneck station is a mirage
4. Increasing the capacity of a bottleneck increases the capacity of
the whole system
Queuing Theory
Queuing Theory is the study of waiting lines.
Waiting lines are common situations with an Arrival Queue and a Service Process:
- Airport - Landing airplanes – Air Control Service
- Plant - Entering batches into production workstations – Workstation Process Capacity
- Gas Stations – Cars – Gas bomb
- Supermarkets – Shoppers – Checkout Clerks
- Hospital – Patients – Doctors
- Call center – Phone Calls – Operators
- Bank – Customers – Teller
- Maintenance – Machines – Maintenance Operators
- Harbour – Ships - Docks

Characteristics of a Waiting-Line System:

◦ Arrival Characteristics: Population size (limited or unlimited), behaviour (Scheduled or random), statistical distribution (typical a Poisson)
◦ Waiting Line Characteristics: limited vs unlimited length, disciplines of users (do not switch lines, wait in queue, balking or impatience,
reneging), Typically First in First Out but other discipline is possible.
◦ Service Characteristics: Design, statistical distribution of service times, Single Channel vs Multiple Channels and Single Phase vs Multiple Phases.
Queuing Costs: Minimizing the total cost: Service +
Waiting Time Costs

COST TOTAL COST

SERVICE COSTS

WAITING TIME COST

OPTIMUM CAPACITY SERVICE CAPACITY

14
Single Channel
1. Arrivals are served on a FIFO basis and every arrival waits to be served regardless of the
length of the queue
2. Arrivals are independent of preceding arrivals but the average number of arrivals does not
change over time
3. Arrivals are described by Poisson distribution and infinite population
4. Service times vary from customer to customer, but average rate is known
5. Service rate is faster than arrival rate
 Variables: Users in the queue and system
Lq: Number of users in the
queue

λ= User being served at a speed

Arrival equal to µ units
Rate

Ls: Number of users in

the system
 Variables: Time in the queue and system
Wq: Time spent in the queue

λ=
Arrival Service Time
Rate

Ws: Time spent in the system

Variables
λ = Mean number of arrivals per time period
µ = Mean number of units served per time period. µ must be larger than λ.
Ls = Average number of units (customers) in the system (waiting and being served) = λ / (µ-λ)
Ws = Average time a unit spends in the system (waiting and being served) = 1/(µ-λ)
Lq = Average number of units (customers) in the queue = λ x Ls /µ = λ2 /µ (µ-λ)
Wq = Average time a unit spends waiting in the queue = Lq / λ = λ /µ (µ-λ)
ρ = Utilization factor = λ /µ
P0 = Probability of the service unit being idle = 1 – ρ
Pn>k = Probability of more than k units in the system, where n is the number of units in the
system = (λ /µ)k+1
Multiple Channel (s = Number of servers)
SYSTEM

EXIT
SERVER

ARRIVALS EXIT
QUEUE SERVER

EXIT
SERVER

19
 Multiple Channel (s = Number of servers)
Lq: Number of users in the queue

User being served

λ=
Arrival
User being served
Rate

User being served

Ls: Number of users in the system

Variables
λ = Mean number of arrivals per time period
µ = Mean number of units served per time period
Ls = Average number of units (customers) in the system (waiting and being served) = Lq + λ /µ
Ws = Average time a unit spends in the system (waiting and being served) = Wq + 1 /µ
Lq = From Table
Wq = Average time a unit spends waiting in the queue = Lq / λ
ρ = Utilization factor = λ /sµ
Lq for s
Example of Queuing Theory Exercise
Customers who arrive at the faculty photocopy service place their orders on a computer and then
approach the service window to pick up their order. At present, there is a single customer service
unit operation run by two employees, one of whom makes the photocopies while the other delivers
the photocopies and charges the client. The number of customers arriving at a given interval follows
a Poisson distribution with an average of 24 per hour. The average time it takes to serve a customer is
distributed exponentially, with an average of 1.25 minutes. Instead, it is proposed to use an operation
with two service units and a single queue. An employee sits in each unit and makes the photocopies
and charges the customers who arrive there. The service time is exponential, with an average of 2
minutes for each unit. Calculate for the two options (one unit and two units) and decide if it is worth
making the decision to change the operation to two windows with the information available:
— The average number of customers waiting for service and the average waiting time for the service.
— The average time spent by a customer in the system and the average number of customers in the
system.
Example of Queuing Theory Exercise
One Server: M/M/1 with two phases

Photocopy Photocopy
Customers arriving to service service Phase 2
one server Phase 1

1 worker 1 worker
makes charges and
photocopie delivers the
s photocopies
λ = 24 customers per
hour
µ = 48 customers per
hour
Example of Queuing Theory Exercise
One Server: M/M/2 with one phase. λ cannot be larger than s (number of servers) x µ. λ =24;
Photocopy
Capacity of the system= 2x30=60/h. service

1 worker makes
photocopies,
Customers arriving to one delivers and
queue but two servers charges

Photocopy service

1 worker making
photocopies, charges
λ = 24 customers per hour and delivers the
µ = 60/2= 30 customers per photocopies
hour
Example of Queuing Theory Exercise
M/M/1
λ = Mean number of arrivals per time period = 24
µ = Mean number of units served per time period. µ must be larger than λ = 48
Ls = Average number of units (customers) in the system (waiting and being served) = λ / (µ-λ) = 24/(48-
24) = 1 customer.
Ws = Average time a unit spends in the system (waiting and being served) = 1/(µ-λ) = 1/(48-24) =
1/24=0.0416hours = 2.5 minutes
Lq = Average number of units (customers) in the queue = λ x Ls /µ = λ2 /µ (µ-λ) = 24x1/48 =0.5
customers
Wq = Average time a unit spends waiting in the queue = Lq / λ = λ /µ (µ-λ) = 0.5/24 = 0.0208333 hours=
1.25mins.
Example of Queuing Theory Exercise
M/M/2
λ = Mean number of arrivals per time period = 24 customers per hour
µ = Mean number of units served per time period = 30 customers per hour
Ls = Average number of units (customers) in the system (waiting and being served) = Lq + λ /µ =
= 0.1523+24/30 = 0.9523 customers in the system
Ws = Average time a unit spends in the system (waiting and being served) = Wq + 1 /µ =
0.006345 + 1/30 = 0.039678 hours = 2.38 mins
Lq = From Table = λ /µ = 24/30 = 0.8 Lq = 0.1523 customers
Wq = Average time a unit spends waiting in the queue = Lq / λ = 0.1523/24=0.006345 hours =
0.38 mins.
ρ = Utilization factor = λ /sµ
Example of Queuing Theory Exercise
M/M/1 M/M/2
Ls 1 0.9523
Lq 0.5 0.1523
Ws 2.5 minutes 2.38 minutes
Wq 1.25 minutes 0.38 minutes

With the new model (M/M/2) the costs of the system increased from 10 euros / server/hour to 15
euros per server and hour. The firm has calculated that the waiting cost in the line per customer is 20
€/ customer/ hour.
Cost of M/M/1 = Cost of the Service + the Cost of the Waiting Times = 10 euros/hour + ( 1.25
minutes x 24 (arrival rate) x 20€/60mins )= 10/hour + 10/hour = 20 euros /hour
Cost of M/M/2 = Cost of the Service + the Cost of the Waiting Times = 15 x 2 (s) + ( 0.38x24x20€/60
= 30 euros/hour + 3.04 euros/hour )= 33.04 euros / hour
Location Decisions
 Long-term decisions
 Difficult to reverse
 Affect fixed & variable costs
 Transportation cost
 As much as 25% of product price

 Other costs: Taxes, wages, rent etc.

Objective: Maximize benefit of location to firm

 Location Decisions
 Revenue varies little between locations  Costs vary little between market areas
COST FOCUS REVENUE FOCUS
 Location is a major cost factor  Location is a major revenue factor
 Affects shipping & production costs (e.g., labor)  Affects amount of customer contact
 Costs vary greatly between locations  Affects volume of business
Location Decisions
Location Methods
 Factor Rating Method
 List relevant factors
 Assign importance weight to each factor (such as 0 – 1)
 Develop scale for each factor (such as 1 – 100)
 Score each location using factor scale
 Multiply scores by weights for each factor & total

 Select location with maximum total score

Location Methods
 Location Break Even Analysis
 Method of cost-volume analysis used for industrial locations
 Steps
 Determine fixed & variable costs for each location

 Plot total cost for each location (Cost on vertical axis, Annual Volume on horizontal axis)
 Select location with lowest total cost for expected production volume
 Must be above break-even
Location Methods
 Center of Gravity:
 Place existing locations on a coordinate grid
 Grid has arbitrary origin & scale
 Maintains relative distances
 Calculate X & Y coordinates for ‘center of gravity’
 Gives location of distribution center
 Minimizes transportation cost
Location Methods
X Coordinate
dix = x coordinate of
 d ix Wi
i location i
Cx 
 Wi Wi = Volume of
i
goods moved to or from
Y Coordinate location i
 d iy Wi
i
diy = y coordinate of
Cy 
W
i
location i
i